Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- README.md +84 -0
- added_tokens.json +59 -0
- all_results.json +22 -0
- checkpoint-48000/added_tokens.json +59 -0
- checkpoint-48000/config.json +59 -0
- checkpoint-48000/generation_config.json +12 -0
- checkpoint-48000/optimizer.pt +3 -0
- checkpoint-48000/pytorch_model.bin +3 -0
- checkpoint-48000/rng_state_0.pth +3 -0
- checkpoint-48000/rng_state_1.pth +3 -0
- checkpoint-48000/rng_state_2.pth +3 -0
- checkpoint-48000/rng_state_3.pth +3 -0
- checkpoint-48000/scheduler.pt +3 -0
- checkpoint-48000/sentencepiece.bpe.model +3 -0
- checkpoint-48000/special_tokens_map.json +63 -0
- checkpoint-48000/tokenizer.json +3 -0
- checkpoint-48000/tokenizer_config.json +528 -0
- checkpoint-48000/trainer_state.json +715 -0
- checkpoint-48000/training_args.bin +3 -0
- config.json +59 -0
- eval_results.json +10 -0
- generated_predictions.txt +0 -0
- generation_config.json +12 -0
- predict_results.json +9 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +63 -0
- tokenizer.json +3 -0
- tokenizer_config.json +528 -0
- train_results.json +8 -0
- trainer_state.json +956 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-48000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ko
|
| 4 |
+
- en
|
| 5 |
+
base_model: facebook/mbart-large-50-many-to-many-mmt
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
metrics:
|
| 9 |
+
- bleu
|
| 10 |
+
model-index:
|
| 11 |
+
- name: ko-en_mbartLarge_exp20p
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# ko-en_mbartLarge_exp20p
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 1.1451
|
| 23 |
+
- Bleu: 28.9507
|
| 24 |
+
- Gen Len: 18.6702
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 5e-05
|
| 44 |
+
- train_batch_size: 4
|
| 45 |
+
- eval_batch_size: 4
|
| 46 |
+
- seed: 42
|
| 47 |
+
- distributed_type: multi-GPU
|
| 48 |
+
- num_devices: 4
|
| 49 |
+
- gradient_accumulation_steps: 2
|
| 50 |
+
- total_train_batch_size: 32
|
| 51 |
+
- total_eval_batch_size: 16
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: cosine_with_restarts
|
| 54 |
+
- lr_scheduler_warmup_steps: 2000
|
| 55 |
+
- num_epochs: 40
|
| 56 |
+
|
| 57 |
+
### Training results
|
| 58 |
+
|
| 59 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 60 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
|
| 61 |
+
| 1.4008 | 0.46 | 4000 | 1.3739 | 22.7174 | 18.7094 |
|
| 62 |
+
| 1.2847 | 0.93 | 8000 | 1.2652 | 24.8557 | 18.7254 |
|
| 63 |
+
| 1.2009 | 1.39 | 12000 | 1.2082 | 26.2074 | 18.7513 |
|
| 64 |
+
| 1.1686 | 1.86 | 16000 | 1.1841 | 26.304 | 19.161 |
|
| 65 |
+
| 1.0205 | 2.32 | 20000 | 1.1441 | 27.8937 | 18.6638 |
|
| 66 |
+
| 1.0217 | 2.78 | 24000 | 1.1301 | 28.4149 | 18.6666 |
|
| 67 |
+
| 0.8876 | 3.25 | 28000 | 1.1270 | 28.5803 | 18.6229 |
|
| 68 |
+
| 0.9024 | 3.71 | 32000 | 1.1181 | 28.852 | 18.7813 |
|
| 69 |
+
| 0.7927 | 4.18 | 36000 | 1.1393 | 28.3975 | 18.4863 |
|
| 70 |
+
| 0.8174 | 4.64 | 40000 | 1.1249 | 28.6313 | 18.3916 |
|
| 71 |
+
| 0.7434 | 5.11 | 44000 | 1.1696 | 28.2898 | 18.7739 |
|
| 72 |
+
| 0.7416 | 5.57 | 48000 | 1.1451 | 28.9507 | 18.6744 |
|
| 73 |
+
| 0.689 | 6.03 | 52000 | 1.1759 | 28.3532 | 18.4481 |
|
| 74 |
+
| 0.7238 | 6.5 | 56000 | 1.1825 | 28.3827 | 18.7038 |
|
| 75 |
+
| 0.7238 | 6.96 | 60000 | 1.1676 | 28.8248 | 18.5073 |
|
| 76 |
+
| 0.657 | 7.43 | 64000 | 1.2514 | 27.4378 | 18.4196 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- Transformers 4.34.0
|
| 82 |
+
- Pytorch 2.1.0+cu121
|
| 83 |
+
- Datasets 2.14.5
|
| 84 |
+
- Tokenizers 0.14.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<mask>": 250053,
|
| 4 |
+
"<pad>": 1,
|
| 5 |
+
"<s>": 0,
|
| 6 |
+
"<unk>": 3,
|
| 7 |
+
"af_ZA": 250026,
|
| 8 |
+
"ar_AR": 250001,
|
| 9 |
+
"az_AZ": 250027,
|
| 10 |
+
"bn_IN": 250028,
|
| 11 |
+
"cs_CZ": 250002,
|
| 12 |
+
"de_DE": 250003,
|
| 13 |
+
"en_XX": 250004,
|
| 14 |
+
"es_XX": 250005,
|
| 15 |
+
"et_EE": 250006,
|
| 16 |
+
"fa_IR": 250029,
|
| 17 |
+
"fi_FI": 250007,
|
| 18 |
+
"fr_XX": 250008,
|
| 19 |
+
"gl_ES": 250051,
|
| 20 |
+
"gu_IN": 250009,
|
| 21 |
+
"he_IL": 250030,
|
| 22 |
+
"hi_IN": 250010,
|
| 23 |
+
"hr_HR": 250031,
|
| 24 |
+
"id_ID": 250032,
|
| 25 |
+
"it_IT": 250011,
|
| 26 |
+
"ja_XX": 250012,
|
| 27 |
+
"ka_GE": 250033,
|
| 28 |
+
"kk_KZ": 250013,
|
| 29 |
+
"km_KH": 250034,
|
| 30 |
+
"ko_KR": 250014,
|
| 31 |
+
"lt_LT": 250015,
|
| 32 |
+
"lv_LV": 250016,
|
| 33 |
+
"mk_MK": 250035,
|
| 34 |
+
"ml_IN": 250036,
|
| 35 |
+
"mn_MN": 250037,
|
| 36 |
+
"mr_IN": 250038,
|
| 37 |
+
"my_MM": 250017,
|
| 38 |
+
"ne_NP": 250018,
|
| 39 |
+
"nl_XX": 250019,
|
| 40 |
+
"pl_PL": 250039,
|
| 41 |
+
"ps_AF": 250040,
|
| 42 |
+
"pt_XX": 250041,
|
| 43 |
+
"ro_RO": 250020,
|
| 44 |
+
"ru_RU": 250021,
|
| 45 |
+
"si_LK": 250022,
|
| 46 |
+
"sl_SI": 250052,
|
| 47 |
+
"sv_SE": 250042,
|
| 48 |
+
"sw_KE": 250043,
|
| 49 |
+
"ta_IN": 250044,
|
| 50 |
+
"te_IN": 250045,
|
| 51 |
+
"th_TH": 250046,
|
| 52 |
+
"tl_XX": 250047,
|
| 53 |
+
"tr_TR": 250023,
|
| 54 |
+
"uk_UA": 250048,
|
| 55 |
+
"ur_PK": 250049,
|
| 56 |
+
"vi_VN": 250024,
|
| 57 |
+
"xh_ZA": 250050,
|
| 58 |
+
"zh_CN": 250025
|
| 59 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.43,
|
| 3 |
+
"eval_bleu": 28.9507,
|
| 4 |
+
"eval_gen_len": 18.6702,
|
| 5 |
+
"eval_loss": 1.1450848579406738,
|
| 6 |
+
"eval_runtime": 1147.1444,
|
| 7 |
+
"eval_samples": 17237,
|
| 8 |
+
"eval_samples_per_second": 15.026,
|
| 9 |
+
"eval_steps_per_second": 0.94,
|
| 10 |
+
"predict_bleu": 29.0444,
|
| 11 |
+
"predict_gen_len": 18.6292,
|
| 12 |
+
"predict_loss": 1.1273916959762573,
|
| 13 |
+
"predict_runtime": 1153.4441,
|
| 14 |
+
"predict_samples": 17237,
|
| 15 |
+
"predict_samples_per_second": 14.944,
|
| 16 |
+
"predict_steps_per_second": 0.935,
|
| 17 |
+
"train_loss": 0.9512434206008911,
|
| 18 |
+
"train_runtime": 65044.6534,
|
| 19 |
+
"train_samples": 137900,
|
| 20 |
+
"train_samples_per_second": 84.803,
|
| 21 |
+
"train_steps_per_second": 5.3
|
| 22 |
+
}
|
checkpoint-48000/added_tokens.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<mask>": 250053,
|
| 4 |
+
"<pad>": 1,
|
| 5 |
+
"<s>": 0,
|
| 6 |
+
"<unk>": 3,
|
| 7 |
+
"af_ZA": 250026,
|
| 8 |
+
"ar_AR": 250001,
|
| 9 |
+
"az_AZ": 250027,
|
| 10 |
+
"bn_IN": 250028,
|
| 11 |
+
"cs_CZ": 250002,
|
| 12 |
+
"de_DE": 250003,
|
| 13 |
+
"en_XX": 250004,
|
| 14 |
+
"es_XX": 250005,
|
| 15 |
+
"et_EE": 250006,
|
| 16 |
+
"fa_IR": 250029,
|
| 17 |
+
"fi_FI": 250007,
|
| 18 |
+
"fr_XX": 250008,
|
| 19 |
+
"gl_ES": 250051,
|
| 20 |
+
"gu_IN": 250009,
|
| 21 |
+
"he_IL": 250030,
|
| 22 |
+
"hi_IN": 250010,
|
| 23 |
+
"hr_HR": 250031,
|
| 24 |
+
"id_ID": 250032,
|
| 25 |
+
"it_IT": 250011,
|
| 26 |
+
"ja_XX": 250012,
|
| 27 |
+
"ka_GE": 250033,
|
| 28 |
+
"kk_KZ": 250013,
|
| 29 |
+
"km_KH": 250034,
|
| 30 |
+
"ko_KR": 250014,
|
| 31 |
+
"lt_LT": 250015,
|
| 32 |
+
"lv_LV": 250016,
|
| 33 |
+
"mk_MK": 250035,
|
| 34 |
+
"ml_IN": 250036,
|
| 35 |
+
"mn_MN": 250037,
|
| 36 |
+
"mr_IN": 250038,
|
| 37 |
+
"my_MM": 250017,
|
| 38 |
+
"ne_NP": 250018,
|
| 39 |
+
"nl_XX": 250019,
|
| 40 |
+
"pl_PL": 250039,
|
| 41 |
+
"ps_AF": 250040,
|
| 42 |
+
"pt_XX": 250041,
|
| 43 |
+
"ro_RO": 250020,
|
| 44 |
+
"ru_RU": 250021,
|
| 45 |
+
"si_LK": 250022,
|
| 46 |
+
"sl_SI": 250052,
|
| 47 |
+
"sv_SE": 250042,
|
| 48 |
+
"sw_KE": 250043,
|
| 49 |
+
"ta_IN": 250044,
|
| 50 |
+
"te_IN": 250045,
|
| 51 |
+
"th_TH": 250046,
|
| 52 |
+
"tl_XX": 250047,
|
| 53 |
+
"tr_TR": 250023,
|
| 54 |
+
"uk_UA": 250048,
|
| 55 |
+
"ur_PK": 250049,
|
| 56 |
+
"vi_VN": 250024,
|
| 57 |
+
"xh_ZA": 250050,
|
| 58 |
+
"zh_CN": 250025
|
| 59 |
+
}
|
checkpoint-48000/config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MBartForConditionalGeneration"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 12,
|
| 20 |
+
"decoder_start_token_id": 2,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"early_stopping": true,
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 4096,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 2,
|
| 28 |
+
"forced_bos_token_id": 250004,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"max_length": 200,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"model_type": "mbart",
|
| 46 |
+
"normalize_before": true,
|
| 47 |
+
"normalize_embedding": true,
|
| 48 |
+
"num_beams": 5,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"scale_embedding": true,
|
| 53 |
+
"static_position_embeddings": false,
|
| 54 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.34.0",
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 250054
|
| 59 |
+
}
|
checkpoint-48000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 250004,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"max_length": 200,
|
| 9 |
+
"num_beams": 5,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.34.0"
|
| 12 |
+
}
|
checkpoint-48000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c323636882d923d1a31ce5e94aabdc39eeb9741d3ab0f6d0f73083aa37bea4bf
|
| 3 |
+
size 4887473903
|
checkpoint-48000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e91e95b6eaedd719e81e3ad5a3700a211291f4d5ea26bb0bcaa45c9a0d85c491
|
| 3 |
+
size 2444694490
|
checkpoint-48000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3892475760043803792f4cdb3ad77ac853f10bb1930685c52cac58bdb46e04bd
|
| 3 |
+
size 15024
|
checkpoint-48000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bfe2f109f8d63ae3f2bc0151c7f474f416a624d91669e5dfe50abcd13530e0e
|
| 3 |
+
size 15024
|
checkpoint-48000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0004b1a01437d767adfdffac62b24027ba9e7ae889eb01b7af0cf1edbc98193
|
| 3 |
+
size 15024
|
checkpoint-48000/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e86d8e2dcb7bb631365af170165b4a2ceca23a4f1bae4b63583d81ba9bffa94
|
| 3 |
+
size 15024
|
checkpoint-48000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62d9291965f6a024aca9c79e677d0a3f40ee4709f787d040aa50665de89cb257
|
| 3 |
+
size 1064
|
checkpoint-48000/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-48000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"ar_AR",
|
| 4 |
+
"cs_CZ",
|
| 5 |
+
"de_DE",
|
| 6 |
+
"en_XX",
|
| 7 |
+
"es_XX",
|
| 8 |
+
"et_EE",
|
| 9 |
+
"fi_FI",
|
| 10 |
+
"fr_XX",
|
| 11 |
+
"gu_IN",
|
| 12 |
+
"hi_IN",
|
| 13 |
+
"it_IT",
|
| 14 |
+
"ja_XX",
|
| 15 |
+
"kk_KZ",
|
| 16 |
+
"ko_KR",
|
| 17 |
+
"lt_LT",
|
| 18 |
+
"lv_LV",
|
| 19 |
+
"my_MM",
|
| 20 |
+
"ne_NP",
|
| 21 |
+
"nl_XX",
|
| 22 |
+
"ro_RO",
|
| 23 |
+
"ru_RU",
|
| 24 |
+
"si_LK",
|
| 25 |
+
"tr_TR",
|
| 26 |
+
"vi_VN",
|
| 27 |
+
"zh_CN",
|
| 28 |
+
"af_ZA",
|
| 29 |
+
"az_AZ",
|
| 30 |
+
"bn_IN",
|
| 31 |
+
"fa_IR",
|
| 32 |
+
"he_IL",
|
| 33 |
+
"hr_HR",
|
| 34 |
+
"id_ID",
|
| 35 |
+
"ka_GE",
|
| 36 |
+
"km_KH",
|
| 37 |
+
"mk_MK",
|
| 38 |
+
"ml_IN",
|
| 39 |
+
"mn_MN",
|
| 40 |
+
"mr_IN",
|
| 41 |
+
"pl_PL",
|
| 42 |
+
"ps_AF",
|
| 43 |
+
"pt_XX",
|
| 44 |
+
"sv_SE",
|
| 45 |
+
"sw_KE",
|
| 46 |
+
"ta_IN",
|
| 47 |
+
"te_IN",
|
| 48 |
+
"th_TH",
|
| 49 |
+
"tl_XX",
|
| 50 |
+
"uk_UA",
|
| 51 |
+
"ur_PK",
|
| 52 |
+
"xh_ZA",
|
| 53 |
+
"gl_ES",
|
| 54 |
+
"sl_SI"
|
| 55 |
+
],
|
| 56 |
+
"bos_token": "<s>",
|
| 57 |
+
"cls_token": "<s>",
|
| 58 |
+
"eos_token": "</s>",
|
| 59 |
+
"mask_token": "<mask>",
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"sep_token": "</s>",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
checkpoint-48000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
|
| 3 |
+
size 17109939
|
checkpoint-48000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "ar_AR",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": true,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": true,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": true,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": true,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": true,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": true,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": true,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": true,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": true,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": true,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": true,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": true,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": true,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": true,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": true,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": true,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": true,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": true,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": true,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": true,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": true,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": true,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": true,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": true,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": true,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": true,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": true,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": true,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": true,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": true,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": true,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": true,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": true,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": true,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": true,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": true,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": true,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": true,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": true,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": true,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": true,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": true,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": true,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "af_ZA",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": true,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "az_AZ",
|
| 245 |
+
"lstrip": true,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": true,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"250028": {
|
| 252 |
+
"content": "bn_IN",
|
| 253 |
+
"lstrip": true,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": true,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"250029": {
|
| 260 |
+
"content": "fa_IR",
|
| 261 |
+
"lstrip": true,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": true,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"250030": {
|
| 268 |
+
"content": "he_IL",
|
| 269 |
+
"lstrip": true,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": true,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"250031": {
|
| 276 |
+
"content": "hr_HR",
|
| 277 |
+
"lstrip": true,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": true,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"250032": {
|
| 284 |
+
"content": "id_ID",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": true,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"250033": {
|
| 292 |
+
"content": "ka_GE",
|
| 293 |
+
"lstrip": true,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": true,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"250034": {
|
| 300 |
+
"content": "km_KH",
|
| 301 |
+
"lstrip": true,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": true,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"250035": {
|
| 308 |
+
"content": "mk_MK",
|
| 309 |
+
"lstrip": true,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": true,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"250036": {
|
| 316 |
+
"content": "ml_IN",
|
| 317 |
+
"lstrip": true,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": true,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"250037": {
|
| 324 |
+
"content": "mn_MN",
|
| 325 |
+
"lstrip": true,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": true,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"250038": {
|
| 332 |
+
"content": "mr_IN",
|
| 333 |
+
"lstrip": true,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": true,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"250039": {
|
| 340 |
+
"content": "pl_PL",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": true,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"250040": {
|
| 348 |
+
"content": "ps_AF",
|
| 349 |
+
"lstrip": true,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": true,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"250041": {
|
| 356 |
+
"content": "pt_XX",
|
| 357 |
+
"lstrip": true,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": true,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"250042": {
|
| 364 |
+
"content": "sv_SE",
|
| 365 |
+
"lstrip": true,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": true,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"250043": {
|
| 372 |
+
"content": "sw_KE",
|
| 373 |
+
"lstrip": true,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": true,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"250044": {
|
| 380 |
+
"content": "ta_IN",
|
| 381 |
+
"lstrip": true,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": true,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"250045": {
|
| 388 |
+
"content": "te_IN",
|
| 389 |
+
"lstrip": true,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": true,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"250046": {
|
| 396 |
+
"content": "th_TH",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": true,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"250047": {
|
| 404 |
+
"content": "tl_XX",
|
| 405 |
+
"lstrip": true,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": true,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"250048": {
|
| 412 |
+
"content": "uk_UA",
|
| 413 |
+
"lstrip": true,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": true,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"250049": {
|
| 420 |
+
"content": "ur_PK",
|
| 421 |
+
"lstrip": true,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": true,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"250050": {
|
| 428 |
+
"content": "xh_ZA",
|
| 429 |
+
"lstrip": true,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": true,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"250051": {
|
| 436 |
+
"content": "gl_ES",
|
| 437 |
+
"lstrip": true,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": true,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"250052": {
|
| 444 |
+
"content": "sl_SI",
|
| 445 |
+
"lstrip": true,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": true,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"250053": {
|
| 452 |
+
"content": "<mask>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
}
|
| 459 |
+
},
|
| 460 |
+
"additional_special_tokens": [
|
| 461 |
+
"ar_AR",
|
| 462 |
+
"cs_CZ",
|
| 463 |
+
"de_DE",
|
| 464 |
+
"en_XX",
|
| 465 |
+
"es_XX",
|
| 466 |
+
"et_EE",
|
| 467 |
+
"fi_FI",
|
| 468 |
+
"fr_XX",
|
| 469 |
+
"gu_IN",
|
| 470 |
+
"hi_IN",
|
| 471 |
+
"it_IT",
|
| 472 |
+
"ja_XX",
|
| 473 |
+
"kk_KZ",
|
| 474 |
+
"ko_KR",
|
| 475 |
+
"lt_LT",
|
| 476 |
+
"lv_LV",
|
| 477 |
+
"my_MM",
|
| 478 |
+
"ne_NP",
|
| 479 |
+
"nl_XX",
|
| 480 |
+
"ro_RO",
|
| 481 |
+
"ru_RU",
|
| 482 |
+
"si_LK",
|
| 483 |
+
"tr_TR",
|
| 484 |
+
"vi_VN",
|
| 485 |
+
"zh_CN",
|
| 486 |
+
"af_ZA",
|
| 487 |
+
"az_AZ",
|
| 488 |
+
"bn_IN",
|
| 489 |
+
"fa_IR",
|
| 490 |
+
"he_IL",
|
| 491 |
+
"hr_HR",
|
| 492 |
+
"id_ID",
|
| 493 |
+
"ka_GE",
|
| 494 |
+
"km_KH",
|
| 495 |
+
"mk_MK",
|
| 496 |
+
"ml_IN",
|
| 497 |
+
"mn_MN",
|
| 498 |
+
"mr_IN",
|
| 499 |
+
"pl_PL",
|
| 500 |
+
"ps_AF",
|
| 501 |
+
"pt_XX",
|
| 502 |
+
"sv_SE",
|
| 503 |
+
"sw_KE",
|
| 504 |
+
"ta_IN",
|
| 505 |
+
"te_IN",
|
| 506 |
+
"th_TH",
|
| 507 |
+
"tl_XX",
|
| 508 |
+
"uk_UA",
|
| 509 |
+
"ur_PK",
|
| 510 |
+
"xh_ZA",
|
| 511 |
+
"gl_ES",
|
| 512 |
+
"sl_SI"
|
| 513 |
+
],
|
| 514 |
+
"bos_token": "<s>",
|
| 515 |
+
"clean_up_tokenization_spaces": true,
|
| 516 |
+
"cls_token": "<s>",
|
| 517 |
+
"eos_token": "</s>",
|
| 518 |
+
"language_codes": "ML50",
|
| 519 |
+
"mask_token": "<mask>",
|
| 520 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 521 |
+
"pad_token": "<pad>",
|
| 522 |
+
"sep_token": "</s>",
|
| 523 |
+
"sp_model_kwargs": {},
|
| 524 |
+
"src_lang": "ko_KR",
|
| 525 |
+
"tgt_lang": "en_XX",
|
| 526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 527 |
+
"unk_token": "<unk>"
|
| 528 |
+
}
|
checkpoint-48000/trainer_state.json
ADDED
|
@@ -0,0 +1,715 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 28.9507,
|
| 3 |
+
"best_model_checkpoint": "./ko-en_mbartLarge_exp20p/checkpoint-48000",
|
| 4 |
+
"epoch": 5.569091541942221,
|
| 5 |
+
"eval_steps": 4000,
|
| 6 |
+
"global_step": 48000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 1.25e-05,
|
| 14 |
+
"loss": 1.9162,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.12,
|
| 19 |
+
"learning_rate": 2.5e-05,
|
| 20 |
+
"loss": 1.8085,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.17,
|
| 25 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 26 |
+
"loss": 1.6542,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.23,
|
| 31 |
+
"learning_rate": 5e-05,
|
| 32 |
+
"loss": 1.543,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 4.999973747616574e-05,
|
| 38 |
+
"loss": 1.4815,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.35,
|
| 43 |
+
"learning_rate": 4.999894991017647e-05,
|
| 44 |
+
"loss": 1.4549,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.41,
|
| 49 |
+
"learning_rate": 4.999763731857256e-05,
|
| 50 |
+
"loss": 1.4233,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.46,
|
| 55 |
+
"learning_rate": 4.9995799728920956e-05,
|
| 56 |
+
"loss": 1.4008,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.46,
|
| 61 |
+
"eval_bleu": 22.7174,
|
| 62 |
+
"eval_gen_len": 18.7094,
|
| 63 |
+
"eval_loss": 1.3739314079284668,
|
| 64 |
+
"eval_runtime": 1153.4033,
|
| 65 |
+
"eval_samples_per_second": 14.944,
|
| 66 |
+
"eval_steps_per_second": 0.935,
|
| 67 |
+
"step": 4000
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.52,
|
| 71 |
+
"learning_rate": 4.9993437179814526e-05,
|
| 72 |
+
"loss": 1.3868,
|
| 73 |
+
"step": 4500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.58,
|
| 77 |
+
"learning_rate": 4.999054972087132e-05,
|
| 78 |
+
"loss": 1.3612,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.64,
|
| 83 |
+
"learning_rate": 4.9987137412733476e-05,
|
| 84 |
+
"loss": 1.3519,
|
| 85 |
+
"step": 5500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.7,
|
| 89 |
+
"learning_rate": 4.998320032706597e-05,
|
| 90 |
+
"loss": 1.351,
|
| 91 |
+
"step": 6000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.75,
|
| 95 |
+
"learning_rate": 4.997873854655512e-05,
|
| 96 |
+
"loss": 1.3285,
|
| 97 |
+
"step": 6500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.81,
|
| 101 |
+
"learning_rate": 4.9973752164906806e-05,
|
| 102 |
+
"loss": 1.2959,
|
| 103 |
+
"step": 7000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.87,
|
| 107 |
+
"learning_rate": 4.9968241286844556e-05,
|
| 108 |
+
"loss": 1.3041,
|
| 109 |
+
"step": 7500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.93,
|
| 113 |
+
"learning_rate": 4.996220602810732e-05,
|
| 114 |
+
"loss": 1.2847,
|
| 115 |
+
"step": 8000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.93,
|
| 119 |
+
"eval_bleu": 24.8557,
|
| 120 |
+
"eval_gen_len": 18.7254,
|
| 121 |
+
"eval_loss": 1.2651795148849487,
|
| 122 |
+
"eval_runtime": 1152.6426,
|
| 123 |
+
"eval_samples_per_second": 14.954,
|
| 124 |
+
"eval_steps_per_second": 0.935,
|
| 125 |
+
"step": 8000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.99,
|
| 129 |
+
"learning_rate": 4.995564651544704e-05,
|
| 130 |
+
"loss": 1.2796,
|
| 131 |
+
"step": 8500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.04,
|
| 135 |
+
"learning_rate": 4.994856288662599e-05,
|
| 136 |
+
"loss": 1.2328,
|
| 137 |
+
"step": 9000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.1,
|
| 141 |
+
"learning_rate": 4.9940955290413884e-05,
|
| 142 |
+
"loss": 1.2225,
|
| 143 |
+
"step": 9500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.16,
|
| 147 |
+
"learning_rate": 4.993282388658474e-05,
|
| 148 |
+
"loss": 1.2061,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.22,
|
| 153 |
+
"learning_rate": 4.992416884591354e-05,
|
| 154 |
+
"loss": 1.2152,
|
| 155 |
+
"step": 10500
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.28,
|
| 159 |
+
"learning_rate": 4.991499035017266e-05,
|
| 160 |
+
"loss": 1.1955,
|
| 161 |
+
"step": 11000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.33,
|
| 165 |
+
"learning_rate": 4.9905288592127996e-05,
|
| 166 |
+
"loss": 1.1855,
|
| 167 |
+
"step": 11500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.39,
|
| 171 |
+
"learning_rate": 4.9895063775534965e-05,
|
| 172 |
+
"loss": 1.2009,
|
| 173 |
+
"step": 12000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.39,
|
| 177 |
+
"eval_bleu": 26.2074,
|
| 178 |
+
"eval_gen_len": 18.7513,
|
| 179 |
+
"eval_loss": 1.20820152759552,
|
| 180 |
+
"eval_runtime": 1153.2349,
|
| 181 |
+
"eval_samples_per_second": 14.947,
|
| 182 |
+
"eval_steps_per_second": 0.935,
|
| 183 |
+
"step": 12000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.45,
|
| 187 |
+
"learning_rate": 4.9884316115134224e-05,
|
| 188 |
+
"loss": 1.1941,
|
| 189 |
+
"step": 12500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.51,
|
| 193 |
+
"learning_rate": 4.987304583664712e-05,
|
| 194 |
+
"loss": 1.1982,
|
| 195 |
+
"step": 13000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.57,
|
| 199 |
+
"learning_rate": 4.986125317677099e-05,
|
| 200 |
+
"loss": 1.188,
|
| 201 |
+
"step": 13500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.62,
|
| 205 |
+
"learning_rate": 4.98489383831742e-05,
|
| 206 |
+
"loss": 1.1825,
|
| 207 |
+
"step": 14000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.68,
|
| 211 |
+
"learning_rate": 4.983610171449087e-05,
|
| 212 |
+
"loss": 1.1948,
|
| 213 |
+
"step": 14500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.74,
|
| 217 |
+
"learning_rate": 4.982274344031553e-05,
|
| 218 |
+
"loss": 1.1476,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.8,
|
| 223 |
+
"learning_rate": 4.980886384119741e-05,
|
| 224 |
+
"loss": 1.1627,
|
| 225 |
+
"step": 15500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.86,
|
| 229 |
+
"learning_rate": 4.979446320863455e-05,
|
| 230 |
+
"loss": 1.1686,
|
| 231 |
+
"step": 16000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.86,
|
| 235 |
+
"eval_bleu": 26.304,
|
| 236 |
+
"eval_gen_len": 19.161,
|
| 237 |
+
"eval_loss": 1.1840966939926147,
|
| 238 |
+
"eval_runtime": 1166.1525,
|
| 239 |
+
"eval_samples_per_second": 14.781,
|
| 240 |
+
"eval_steps_per_second": 0.924,
|
| 241 |
+
"step": 16000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.91,
|
| 245 |
+
"learning_rate": 4.97795418450677e-05,
|
| 246 |
+
"loss": 1.1675,
|
| 247 |
+
"step": 16500
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.97,
|
| 251 |
+
"learning_rate": 4.976410006387394e-05,
|
| 252 |
+
"loss": 1.1585,
|
| 253 |
+
"step": 17000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.03,
|
| 257 |
+
"learning_rate": 4.974813818936012e-05,
|
| 258 |
+
"loss": 1.0833,
|
| 259 |
+
"step": 17500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.09,
|
| 263 |
+
"learning_rate": 4.973165655675605e-05,
|
| 264 |
+
"loss": 1.0134,
|
| 265 |
+
"step": 18000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.15,
|
| 269 |
+
"learning_rate": 4.9714655512207414e-05,
|
| 270 |
+
"loss": 1.0257,
|
| 271 |
+
"step": 18500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.2,
|
| 275 |
+
"learning_rate": 4.969713541276859e-05,
|
| 276 |
+
"loss": 1.0206,
|
| 277 |
+
"step": 19000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.26,
|
| 281 |
+
"learning_rate": 4.967909662639506e-05,
|
| 282 |
+
"loss": 1.0503,
|
| 283 |
+
"step": 19500
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.32,
|
| 287 |
+
"learning_rate": 4.9660539531935746e-05,
|
| 288 |
+
"loss": 1.0205,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.32,
|
| 293 |
+
"eval_bleu": 27.8937,
|
| 294 |
+
"eval_gen_len": 18.6638,
|
| 295 |
+
"eval_loss": 1.144053339958191,
|
| 296 |
+
"eval_runtime": 1147.812,
|
| 297 |
+
"eval_samples_per_second": 15.017,
|
| 298 |
+
"eval_steps_per_second": 0.939,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.38,
|
| 303 |
+
"learning_rate": 4.9641464519125e-05,
|
| 304 |
+
"loss": 1.032,
|
| 305 |
+
"step": 20500
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.44,
|
| 309 |
+
"learning_rate": 4.962187198857447e-05,
|
| 310 |
+
"loss": 1.0121,
|
| 311 |
+
"step": 21000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.49,
|
| 315 |
+
"learning_rate": 4.960176235176465e-05,
|
| 316 |
+
"loss": 1.013,
|
| 317 |
+
"step": 21500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.55,
|
| 321 |
+
"learning_rate": 4.958113603103627e-05,
|
| 322 |
+
"loss": 1.0169,
|
| 323 |
+
"step": 22000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.61,
|
| 327 |
+
"learning_rate": 4.9559993459581375e-05,
|
| 328 |
+
"loss": 1.0315,
|
| 329 |
+
"step": 22500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.67,
|
| 333 |
+
"learning_rate": 4.953833508143429e-05,
|
| 334 |
+
"loss": 1.0431,
|
| 335 |
+
"step": 23000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.73,
|
| 339 |
+
"learning_rate": 4.951616135146226e-05,
|
| 340 |
+
"loss": 1.0398,
|
| 341 |
+
"step": 23500
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 2.78,
|
| 345 |
+
"learning_rate": 4.949347273535588e-05,
|
| 346 |
+
"loss": 1.0217,
|
| 347 |
+
"step": 24000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.78,
|
| 351 |
+
"eval_bleu": 28.4149,
|
| 352 |
+
"eval_gen_len": 18.6666,
|
| 353 |
+
"eval_loss": 1.1301122903823853,
|
| 354 |
+
"eval_runtime": 1149.3346,
|
| 355 |
+
"eval_samples_per_second": 14.997,
|
| 356 |
+
"eval_steps_per_second": 0.938,
|
| 357 |
+
"step": 24000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.84,
|
| 361 |
+
"learning_rate": 4.9470269709619356e-05,
|
| 362 |
+
"loss": 1.0313,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.9,
|
| 367 |
+
"learning_rate": 4.944655276156047e-05,
|
| 368 |
+
"loss": 1.0127,
|
| 369 |
+
"step": 25000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.96,
|
| 373 |
+
"learning_rate": 4.9422322389280354e-05,
|
| 374 |
+
"loss": 1.0183,
|
| 375 |
+
"step": 25500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.02,
|
| 379 |
+
"learning_rate": 4.939757910166303e-05,
|
| 380 |
+
"loss": 0.9723,
|
| 381 |
+
"step": 26000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.07,
|
| 385 |
+
"learning_rate": 4.937232341836471e-05,
|
| 386 |
+
"loss": 0.8652,
|
| 387 |
+
"step": 26500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.13,
|
| 391 |
+
"learning_rate": 4.9346555869802904e-05,
|
| 392 |
+
"loss": 0.8811,
|
| 393 |
+
"step": 27000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.19,
|
| 397 |
+
"learning_rate": 4.9320276997145255e-05,
|
| 398 |
+
"loss": 0.8824,
|
| 399 |
+
"step": 27500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 3.25,
|
| 403 |
+
"learning_rate": 4.929348735229821e-05,
|
| 404 |
+
"loss": 0.8876,
|
| 405 |
+
"step": 28000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 3.25,
|
| 409 |
+
"eval_bleu": 28.5803,
|
| 410 |
+
"eval_gen_len": 18.6229,
|
| 411 |
+
"eval_loss": 1.1270363330841064,
|
| 412 |
+
"eval_runtime": 1153.2981,
|
| 413 |
+
"eval_samples_per_second": 14.946,
|
| 414 |
+
"eval_steps_per_second": 0.935,
|
| 415 |
+
"step": 28000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.31,
|
| 419 |
+
"learning_rate": 4.926618749789539e-05,
|
| 420 |
+
"loss": 0.8734,
|
| 421 |
+
"step": 28500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.36,
|
| 425 |
+
"learning_rate": 4.923837800728578e-05,
|
| 426 |
+
"loss": 0.8772,
|
| 427 |
+
"step": 29000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.42,
|
| 431 |
+
"learning_rate": 4.921005946452171e-05,
|
| 432 |
+
"loss": 0.8821,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 3.48,
|
| 437 |
+
"learning_rate": 4.918123246434658e-05,
|
| 438 |
+
"loss": 0.8774,
|
| 439 |
+
"step": 30000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 3.54,
|
| 443 |
+
"learning_rate": 4.915189761218237e-05,
|
| 444 |
+
"loss": 0.8932,
|
| 445 |
+
"step": 30500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 3.6,
|
| 449 |
+
"learning_rate": 4.912205552411688e-05,
|
| 450 |
+
"loss": 0.8745,
|
| 451 |
+
"step": 31000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.65,
|
| 455 |
+
"learning_rate": 4.9091706826890896e-05,
|
| 456 |
+
"loss": 0.8965,
|
| 457 |
+
"step": 31500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.71,
|
| 461 |
+
"learning_rate": 4.9060852157884893e-05,
|
| 462 |
+
"loss": 0.9024,
|
| 463 |
+
"step": 32000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.71,
|
| 467 |
+
"eval_bleu": 28.852,
|
| 468 |
+
"eval_gen_len": 18.7813,
|
| 469 |
+
"eval_loss": 1.118120551109314,
|
| 470 |
+
"eval_runtime": 1155.1982,
|
| 471 |
+
"eval_samples_per_second": 14.921,
|
| 472 |
+
"eval_steps_per_second": 0.933,
|
| 473 |
+
"step": 32000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.77,
|
| 477 |
+
"learning_rate": 4.9029492165105776e-05,
|
| 478 |
+
"loss": 0.8909,
|
| 479 |
+
"step": 32500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.83,
|
| 483 |
+
"learning_rate": 4.899762750717318e-05,
|
| 484 |
+
"loss": 0.8989,
|
| 485 |
+
"step": 33000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.89,
|
| 489 |
+
"learning_rate": 4.8965258853305676e-05,
|
| 490 |
+
"loss": 0.9054,
|
| 491 |
+
"step": 33500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.94,
|
| 495 |
+
"learning_rate": 4.893238688330673e-05,
|
| 496 |
+
"loss": 0.8852,
|
| 497 |
+
"step": 34000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.0,
|
| 501 |
+
"learning_rate": 4.889901228755036e-05,
|
| 502 |
+
"loss": 0.8879,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.06,
|
| 507 |
+
"learning_rate": 4.886513576696674e-05,
|
| 508 |
+
"loss": 0.7955,
|
| 509 |
+
"step": 35000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.12,
|
| 513 |
+
"learning_rate": 4.8830758033027385e-05,
|
| 514 |
+
"loss": 0.795,
|
| 515 |
+
"step": 35500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.18,
|
| 519 |
+
"learning_rate": 4.879587980773026e-05,
|
| 520 |
+
"loss": 0.7927,
|
| 521 |
+
"step": 36000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.18,
|
| 525 |
+
"eval_bleu": 28.3975,
|
| 526 |
+
"eval_gen_len": 18.4863,
|
| 527 |
+
"eval_loss": 1.1393358707427979,
|
| 528 |
+
"eval_runtime": 1137.7417,
|
| 529 |
+
"eval_samples_per_second": 15.15,
|
| 530 |
+
"eval_steps_per_second": 0.947,
|
| 531 |
+
"step": 36000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.23,
|
| 535 |
+
"learning_rate": 4.8760501823584607e-05,
|
| 536 |
+
"loss": 0.8031,
|
| 537 |
+
"step": 36500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 4.29,
|
| 541 |
+
"learning_rate": 4.8724624823595535e-05,
|
| 542 |
+
"loss": 0.7891,
|
| 543 |
+
"step": 37000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 4.35,
|
| 547 |
+
"learning_rate": 4.8688249561248456e-05,
|
| 548 |
+
"loss": 0.8028,
|
| 549 |
+
"step": 37500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 4.41,
|
| 553 |
+
"learning_rate": 4.8651376800493245e-05,
|
| 554 |
+
"loss": 0.8059,
|
| 555 |
+
"step": 38000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 4.47,
|
| 559 |
+
"learning_rate": 4.861400731572818e-05,
|
| 560 |
+
"loss": 0.8093,
|
| 561 |
+
"step": 38500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 4.52,
|
| 565 |
+
"learning_rate": 4.857614189178369e-05,
|
| 566 |
+
"loss": 0.8044,
|
| 567 |
+
"step": 39000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 4.58,
|
| 571 |
+
"learning_rate": 4.8537781323905896e-05,
|
| 572 |
+
"loss": 0.8098,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 4.64,
|
| 577 |
+
"learning_rate": 4.849892641773984e-05,
|
| 578 |
+
"loss": 0.8174,
|
| 579 |
+
"step": 40000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 4.64,
|
| 583 |
+
"eval_bleu": 28.6313,
|
| 584 |
+
"eval_gen_len": 18.3916,
|
| 585 |
+
"eval_loss": 1.124934434890747,
|
| 586 |
+
"eval_runtime": 1132.4078,
|
| 587 |
+
"eval_samples_per_second": 15.222,
|
| 588 |
+
"eval_steps_per_second": 0.952,
|
| 589 |
+
"step": 40000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 4.7,
|
| 593 |
+
"learning_rate": 4.845957798931265e-05,
|
| 594 |
+
"loss": 0.821,
|
| 595 |
+
"step": 40500
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.76,
|
| 599 |
+
"learning_rate": 4.841973686501636e-05,
|
| 600 |
+
"loss": 0.8104,
|
| 601 |
+
"step": 41000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 4.81,
|
| 605 |
+
"learning_rate": 4.8379403881590544e-05,
|
| 606 |
+
"loss": 0.8155,
|
| 607 |
+
"step": 41500
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 4.87,
|
| 611 |
+
"learning_rate": 4.833857988610474e-05,
|
| 612 |
+
"loss": 0.8213,
|
| 613 |
+
"step": 42000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 4.93,
|
| 617 |
+
"learning_rate": 4.8297265735940714e-05,
|
| 618 |
+
"loss": 0.819,
|
| 619 |
+
"step": 42500
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 4.99,
|
| 623 |
+
"learning_rate": 4.825546229877439e-05,
|
| 624 |
+
"loss": 0.8255,
|
| 625 |
+
"step": 43000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 5.05,
|
| 629 |
+
"learning_rate": 4.8213170452557655e-05,
|
| 630 |
+
"loss": 0.7598,
|
| 631 |
+
"step": 43500
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 5.11,
|
| 635 |
+
"learning_rate": 4.817039108549991e-05,
|
| 636 |
+
"loss": 0.7434,
|
| 637 |
+
"step": 44000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.11,
|
| 641 |
+
"eval_bleu": 28.2898,
|
| 642 |
+
"eval_gen_len": 18.7739,
|
| 643 |
+
"eval_loss": 1.1695784330368042,
|
| 644 |
+
"eval_runtime": 1160.4522,
|
| 645 |
+
"eval_samples_per_second": 14.854,
|
| 646 |
+
"eval_steps_per_second": 0.929,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 5.16,
|
| 651 |
+
"learning_rate": 4.812712509604945e-05,
|
| 652 |
+
"loss": 0.7564,
|
| 653 |
+
"step": 44500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 5.22,
|
| 657 |
+
"learning_rate": 4.808337339287453e-05,
|
| 658 |
+
"loss": 0.7496,
|
| 659 |
+
"step": 45000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 5.28,
|
| 663 |
+
"learning_rate": 4.803913689484436e-05,
|
| 664 |
+
"loss": 0.7381,
|
| 665 |
+
"step": 45500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.34,
|
| 669 |
+
"learning_rate": 4.7994416531009745e-05,
|
| 670 |
+
"loss": 0.7394,
|
| 671 |
+
"step": 46000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 5.4,
|
| 675 |
+
"learning_rate": 4.794921324058358e-05,
|
| 676 |
+
"loss": 0.7468,
|
| 677 |
+
"step": 46500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.45,
|
| 681 |
+
"learning_rate": 4.790352797292116e-05,
|
| 682 |
+
"loss": 0.7448,
|
| 683 |
+
"step": 47000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.51,
|
| 687 |
+
"learning_rate": 4.785736168750022e-05,
|
| 688 |
+
"loss": 0.7339,
|
| 689 |
+
"step": 47500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.57,
|
| 693 |
+
"learning_rate": 4.781071535390078e-05,
|
| 694 |
+
"loss": 0.7416,
|
| 695 |
+
"step": 48000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.57,
|
| 699 |
+
"eval_bleu": 28.9507,
|
| 700 |
+
"eval_gen_len": 18.6744,
|
| 701 |
+
"eval_loss": 1.1450848579406738,
|
| 702 |
+
"eval_runtime": 1148.6011,
|
| 703 |
+
"eval_samples_per_second": 15.007,
|
| 704 |
+
"eval_steps_per_second": 0.939,
|
| 705 |
+
"step": 48000
|
| 706 |
+
}
|
| 707 |
+
],
|
| 708 |
+
"logging_steps": 500,
|
| 709 |
+
"max_steps": 344760,
|
| 710 |
+
"num_train_epochs": 40,
|
| 711 |
+
"save_steps": 4000,
|
| 712 |
+
"total_flos": 1.6644596837352735e+18,
|
| 713 |
+
"trial_name": null,
|
| 714 |
+
"trial_params": null
|
| 715 |
+
}
|
checkpoint-48000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:812228f1e0a1b305a43ac2ec4cf6a158a90cadcdd15b8aec2eea894cd17b40de
|
| 3 |
+
size 4728
|
config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MBartForConditionalGeneration"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 12,
|
| 20 |
+
"decoder_start_token_id": 2,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"early_stopping": true,
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 4096,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 2,
|
| 28 |
+
"forced_bos_token_id": 250004,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"max_length": 200,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"model_type": "mbart",
|
| 46 |
+
"normalize_before": true,
|
| 47 |
+
"normalize_embedding": true,
|
| 48 |
+
"num_beams": 5,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"scale_embedding": true,
|
| 53 |
+
"static_position_embeddings": false,
|
| 54 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.34.0",
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 250054
|
| 59 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.43,
|
| 3 |
+
"eval_bleu": 28.9507,
|
| 4 |
+
"eval_gen_len": 18.6702,
|
| 5 |
+
"eval_loss": 1.1450848579406738,
|
| 6 |
+
"eval_runtime": 1147.1444,
|
| 7 |
+
"eval_samples": 17237,
|
| 8 |
+
"eval_samples_per_second": 15.026,
|
| 9 |
+
"eval_steps_per_second": 0.94
|
| 10 |
+
}
|
generated_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 250004,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"max_length": 200,
|
| 9 |
+
"num_beams": 5,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.34.0"
|
| 12 |
+
}
|
predict_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"predict_bleu": 29.0444,
|
| 3 |
+
"predict_gen_len": 18.6292,
|
| 4 |
+
"predict_loss": 1.1273916959762573,
|
| 5 |
+
"predict_runtime": 1153.4441,
|
| 6 |
+
"predict_samples": 17237,
|
| 7 |
+
"predict_samples_per_second": 14.944,
|
| 8 |
+
"predict_steps_per_second": 0.935
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e91e95b6eaedd719e81e3ad5a3700a211291f4d5ea26bb0bcaa45c9a0d85c491
|
| 3 |
+
size 2444694490
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"ar_AR",
|
| 4 |
+
"cs_CZ",
|
| 5 |
+
"de_DE",
|
| 6 |
+
"en_XX",
|
| 7 |
+
"es_XX",
|
| 8 |
+
"et_EE",
|
| 9 |
+
"fi_FI",
|
| 10 |
+
"fr_XX",
|
| 11 |
+
"gu_IN",
|
| 12 |
+
"hi_IN",
|
| 13 |
+
"it_IT",
|
| 14 |
+
"ja_XX",
|
| 15 |
+
"kk_KZ",
|
| 16 |
+
"ko_KR",
|
| 17 |
+
"lt_LT",
|
| 18 |
+
"lv_LV",
|
| 19 |
+
"my_MM",
|
| 20 |
+
"ne_NP",
|
| 21 |
+
"nl_XX",
|
| 22 |
+
"ro_RO",
|
| 23 |
+
"ru_RU",
|
| 24 |
+
"si_LK",
|
| 25 |
+
"tr_TR",
|
| 26 |
+
"vi_VN",
|
| 27 |
+
"zh_CN",
|
| 28 |
+
"af_ZA",
|
| 29 |
+
"az_AZ",
|
| 30 |
+
"bn_IN",
|
| 31 |
+
"fa_IR",
|
| 32 |
+
"he_IL",
|
| 33 |
+
"hr_HR",
|
| 34 |
+
"id_ID",
|
| 35 |
+
"ka_GE",
|
| 36 |
+
"km_KH",
|
| 37 |
+
"mk_MK",
|
| 38 |
+
"ml_IN",
|
| 39 |
+
"mn_MN",
|
| 40 |
+
"mr_IN",
|
| 41 |
+
"pl_PL",
|
| 42 |
+
"ps_AF",
|
| 43 |
+
"pt_XX",
|
| 44 |
+
"sv_SE",
|
| 45 |
+
"sw_KE",
|
| 46 |
+
"ta_IN",
|
| 47 |
+
"te_IN",
|
| 48 |
+
"th_TH",
|
| 49 |
+
"tl_XX",
|
| 50 |
+
"uk_UA",
|
| 51 |
+
"ur_PK",
|
| 52 |
+
"xh_ZA",
|
| 53 |
+
"gl_ES",
|
| 54 |
+
"sl_SI"
|
| 55 |
+
],
|
| 56 |
+
"bos_token": "<s>",
|
| 57 |
+
"cls_token": "<s>",
|
| 58 |
+
"eos_token": "</s>",
|
| 59 |
+
"mask_token": "<mask>",
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"sep_token": "</s>",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
|
| 3 |
+
size 17109939
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "ar_AR",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": true,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": true,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": true,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": true,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": true,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": true,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": true,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": true,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": true,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": true,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": true,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": true,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": true,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": true,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": true,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": true,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": true,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": true,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": true,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": true,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": true,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": true,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": true,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": true,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": true,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": true,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": true,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": true,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": true,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": true,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": true,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": true,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": true,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": true,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": true,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": true,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": true,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": true,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": true,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": true,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": true,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": true,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": true,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "af_ZA",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": true,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "az_AZ",
|
| 245 |
+
"lstrip": true,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": true,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"250028": {
|
| 252 |
+
"content": "bn_IN",
|
| 253 |
+
"lstrip": true,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": true,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"250029": {
|
| 260 |
+
"content": "fa_IR",
|
| 261 |
+
"lstrip": true,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": true,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"250030": {
|
| 268 |
+
"content": "he_IL",
|
| 269 |
+
"lstrip": true,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": true,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"250031": {
|
| 276 |
+
"content": "hr_HR",
|
| 277 |
+
"lstrip": true,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": true,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"250032": {
|
| 284 |
+
"content": "id_ID",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": true,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"250033": {
|
| 292 |
+
"content": "ka_GE",
|
| 293 |
+
"lstrip": true,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": true,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"250034": {
|
| 300 |
+
"content": "km_KH",
|
| 301 |
+
"lstrip": true,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": true,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"250035": {
|
| 308 |
+
"content": "mk_MK",
|
| 309 |
+
"lstrip": true,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": true,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"250036": {
|
| 316 |
+
"content": "ml_IN",
|
| 317 |
+
"lstrip": true,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": true,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"250037": {
|
| 324 |
+
"content": "mn_MN",
|
| 325 |
+
"lstrip": true,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": true,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"250038": {
|
| 332 |
+
"content": "mr_IN",
|
| 333 |
+
"lstrip": true,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": true,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"250039": {
|
| 340 |
+
"content": "pl_PL",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": true,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"250040": {
|
| 348 |
+
"content": "ps_AF",
|
| 349 |
+
"lstrip": true,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": true,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"250041": {
|
| 356 |
+
"content": "pt_XX",
|
| 357 |
+
"lstrip": true,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": true,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"250042": {
|
| 364 |
+
"content": "sv_SE",
|
| 365 |
+
"lstrip": true,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": true,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"250043": {
|
| 372 |
+
"content": "sw_KE",
|
| 373 |
+
"lstrip": true,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": true,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"250044": {
|
| 380 |
+
"content": "ta_IN",
|
| 381 |
+
"lstrip": true,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": true,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"250045": {
|
| 388 |
+
"content": "te_IN",
|
| 389 |
+
"lstrip": true,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": true,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"250046": {
|
| 396 |
+
"content": "th_TH",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": true,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"250047": {
|
| 404 |
+
"content": "tl_XX",
|
| 405 |
+
"lstrip": true,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": true,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"250048": {
|
| 412 |
+
"content": "uk_UA",
|
| 413 |
+
"lstrip": true,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": true,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"250049": {
|
| 420 |
+
"content": "ur_PK",
|
| 421 |
+
"lstrip": true,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": true,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"250050": {
|
| 428 |
+
"content": "xh_ZA",
|
| 429 |
+
"lstrip": true,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": true,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"250051": {
|
| 436 |
+
"content": "gl_ES",
|
| 437 |
+
"lstrip": true,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": true,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"250052": {
|
| 444 |
+
"content": "sl_SI",
|
| 445 |
+
"lstrip": true,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": true,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"250053": {
|
| 452 |
+
"content": "<mask>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
}
|
| 459 |
+
},
|
| 460 |
+
"additional_special_tokens": [
|
| 461 |
+
"ar_AR",
|
| 462 |
+
"cs_CZ",
|
| 463 |
+
"de_DE",
|
| 464 |
+
"en_XX",
|
| 465 |
+
"es_XX",
|
| 466 |
+
"et_EE",
|
| 467 |
+
"fi_FI",
|
| 468 |
+
"fr_XX",
|
| 469 |
+
"gu_IN",
|
| 470 |
+
"hi_IN",
|
| 471 |
+
"it_IT",
|
| 472 |
+
"ja_XX",
|
| 473 |
+
"kk_KZ",
|
| 474 |
+
"ko_KR",
|
| 475 |
+
"lt_LT",
|
| 476 |
+
"lv_LV",
|
| 477 |
+
"my_MM",
|
| 478 |
+
"ne_NP",
|
| 479 |
+
"nl_XX",
|
| 480 |
+
"ro_RO",
|
| 481 |
+
"ru_RU",
|
| 482 |
+
"si_LK",
|
| 483 |
+
"tr_TR",
|
| 484 |
+
"vi_VN",
|
| 485 |
+
"zh_CN",
|
| 486 |
+
"af_ZA",
|
| 487 |
+
"az_AZ",
|
| 488 |
+
"bn_IN",
|
| 489 |
+
"fa_IR",
|
| 490 |
+
"he_IL",
|
| 491 |
+
"hr_HR",
|
| 492 |
+
"id_ID",
|
| 493 |
+
"ka_GE",
|
| 494 |
+
"km_KH",
|
| 495 |
+
"mk_MK",
|
| 496 |
+
"ml_IN",
|
| 497 |
+
"mn_MN",
|
| 498 |
+
"mr_IN",
|
| 499 |
+
"pl_PL",
|
| 500 |
+
"ps_AF",
|
| 501 |
+
"pt_XX",
|
| 502 |
+
"sv_SE",
|
| 503 |
+
"sw_KE",
|
| 504 |
+
"ta_IN",
|
| 505 |
+
"te_IN",
|
| 506 |
+
"th_TH",
|
| 507 |
+
"tl_XX",
|
| 508 |
+
"uk_UA",
|
| 509 |
+
"ur_PK",
|
| 510 |
+
"xh_ZA",
|
| 511 |
+
"gl_ES",
|
| 512 |
+
"sl_SI"
|
| 513 |
+
],
|
| 514 |
+
"bos_token": "<s>",
|
| 515 |
+
"clean_up_tokenization_spaces": true,
|
| 516 |
+
"cls_token": "<s>",
|
| 517 |
+
"eos_token": "</s>",
|
| 518 |
+
"language_codes": "ML50",
|
| 519 |
+
"mask_token": "<mask>",
|
| 520 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 521 |
+
"pad_token": "<pad>",
|
| 522 |
+
"sep_token": "</s>",
|
| 523 |
+
"sp_model_kwargs": {},
|
| 524 |
+
"src_lang": "ko_KR",
|
| 525 |
+
"tgt_lang": "en_XX",
|
| 526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 527 |
+
"unk_token": "<unk>"
|
| 528 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.43,
|
| 3 |
+
"train_loss": 0.9512434206008911,
|
| 4 |
+
"train_runtime": 65044.6534,
|
| 5 |
+
"train_samples": 137900,
|
| 6 |
+
"train_samples_per_second": 84.803,
|
| 7 |
+
"train_steps_per_second": 5.3
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,956 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 28.9507,
|
| 3 |
+
"best_model_checkpoint": "./ko-en_mbartLarge_exp20p/checkpoint-48000",
|
| 4 |
+
"epoch": 7.425455389256294,
|
| 5 |
+
"eval_steps": 4000,
|
| 6 |
+
"global_step": 64000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 1.25e-05,
|
| 14 |
+
"loss": 1.9162,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.12,
|
| 19 |
+
"learning_rate": 2.5e-05,
|
| 20 |
+
"loss": 1.8085,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.17,
|
| 25 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 26 |
+
"loss": 1.6542,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.23,
|
| 31 |
+
"learning_rate": 5e-05,
|
| 32 |
+
"loss": 1.543,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 4.999973747616574e-05,
|
| 38 |
+
"loss": 1.4815,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.35,
|
| 43 |
+
"learning_rate": 4.999894991017647e-05,
|
| 44 |
+
"loss": 1.4549,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.41,
|
| 49 |
+
"learning_rate": 4.999763731857256e-05,
|
| 50 |
+
"loss": 1.4233,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.46,
|
| 55 |
+
"learning_rate": 4.9995799728920956e-05,
|
| 56 |
+
"loss": 1.4008,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.46,
|
| 61 |
+
"eval_bleu": 22.7174,
|
| 62 |
+
"eval_gen_len": 18.7094,
|
| 63 |
+
"eval_loss": 1.3739314079284668,
|
| 64 |
+
"eval_runtime": 1153.4033,
|
| 65 |
+
"eval_samples_per_second": 14.944,
|
| 66 |
+
"eval_steps_per_second": 0.935,
|
| 67 |
+
"step": 4000
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.52,
|
| 71 |
+
"learning_rate": 4.9993437179814526e-05,
|
| 72 |
+
"loss": 1.3868,
|
| 73 |
+
"step": 4500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.58,
|
| 77 |
+
"learning_rate": 4.999054972087132e-05,
|
| 78 |
+
"loss": 1.3612,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.64,
|
| 83 |
+
"learning_rate": 4.9987137412733476e-05,
|
| 84 |
+
"loss": 1.3519,
|
| 85 |
+
"step": 5500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.7,
|
| 89 |
+
"learning_rate": 4.998320032706597e-05,
|
| 90 |
+
"loss": 1.351,
|
| 91 |
+
"step": 6000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.75,
|
| 95 |
+
"learning_rate": 4.997873854655512e-05,
|
| 96 |
+
"loss": 1.3285,
|
| 97 |
+
"step": 6500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.81,
|
| 101 |
+
"learning_rate": 4.9973752164906806e-05,
|
| 102 |
+
"loss": 1.2959,
|
| 103 |
+
"step": 7000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.87,
|
| 107 |
+
"learning_rate": 4.9968241286844556e-05,
|
| 108 |
+
"loss": 1.3041,
|
| 109 |
+
"step": 7500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.93,
|
| 113 |
+
"learning_rate": 4.996220602810732e-05,
|
| 114 |
+
"loss": 1.2847,
|
| 115 |
+
"step": 8000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.93,
|
| 119 |
+
"eval_bleu": 24.8557,
|
| 120 |
+
"eval_gen_len": 18.7254,
|
| 121 |
+
"eval_loss": 1.2651795148849487,
|
| 122 |
+
"eval_runtime": 1152.6426,
|
| 123 |
+
"eval_samples_per_second": 14.954,
|
| 124 |
+
"eval_steps_per_second": 0.935,
|
| 125 |
+
"step": 8000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.99,
|
| 129 |
+
"learning_rate": 4.995564651544704e-05,
|
| 130 |
+
"loss": 1.2796,
|
| 131 |
+
"step": 8500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.04,
|
| 135 |
+
"learning_rate": 4.994856288662599e-05,
|
| 136 |
+
"loss": 1.2328,
|
| 137 |
+
"step": 9000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.1,
|
| 141 |
+
"learning_rate": 4.9940955290413884e-05,
|
| 142 |
+
"loss": 1.2225,
|
| 143 |
+
"step": 9500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.16,
|
| 147 |
+
"learning_rate": 4.993282388658474e-05,
|
| 148 |
+
"loss": 1.2061,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.22,
|
| 153 |
+
"learning_rate": 4.992416884591354e-05,
|
| 154 |
+
"loss": 1.2152,
|
| 155 |
+
"step": 10500
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.28,
|
| 159 |
+
"learning_rate": 4.991499035017266e-05,
|
| 160 |
+
"loss": 1.1955,
|
| 161 |
+
"step": 11000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.33,
|
| 165 |
+
"learning_rate": 4.9905288592127996e-05,
|
| 166 |
+
"loss": 1.1855,
|
| 167 |
+
"step": 11500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.39,
|
| 171 |
+
"learning_rate": 4.9895063775534965e-05,
|
| 172 |
+
"loss": 1.2009,
|
| 173 |
+
"step": 12000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.39,
|
| 177 |
+
"eval_bleu": 26.2074,
|
| 178 |
+
"eval_gen_len": 18.7513,
|
| 179 |
+
"eval_loss": 1.20820152759552,
|
| 180 |
+
"eval_runtime": 1153.2349,
|
| 181 |
+
"eval_samples_per_second": 14.947,
|
| 182 |
+
"eval_steps_per_second": 0.935,
|
| 183 |
+
"step": 12000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.45,
|
| 187 |
+
"learning_rate": 4.9884316115134224e-05,
|
| 188 |
+
"loss": 1.1941,
|
| 189 |
+
"step": 12500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.51,
|
| 193 |
+
"learning_rate": 4.987304583664712e-05,
|
| 194 |
+
"loss": 1.1982,
|
| 195 |
+
"step": 13000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.57,
|
| 199 |
+
"learning_rate": 4.986125317677099e-05,
|
| 200 |
+
"loss": 1.188,
|
| 201 |
+
"step": 13500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.62,
|
| 205 |
+
"learning_rate": 4.98489383831742e-05,
|
| 206 |
+
"loss": 1.1825,
|
| 207 |
+
"step": 14000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.68,
|
| 211 |
+
"learning_rate": 4.983610171449087e-05,
|
| 212 |
+
"loss": 1.1948,
|
| 213 |
+
"step": 14500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.74,
|
| 217 |
+
"learning_rate": 4.982274344031553e-05,
|
| 218 |
+
"loss": 1.1476,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.8,
|
| 223 |
+
"learning_rate": 4.980886384119741e-05,
|
| 224 |
+
"loss": 1.1627,
|
| 225 |
+
"step": 15500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.86,
|
| 229 |
+
"learning_rate": 4.979446320863455e-05,
|
| 230 |
+
"loss": 1.1686,
|
| 231 |
+
"step": 16000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.86,
|
| 235 |
+
"eval_bleu": 26.304,
|
| 236 |
+
"eval_gen_len": 19.161,
|
| 237 |
+
"eval_loss": 1.1840966939926147,
|
| 238 |
+
"eval_runtime": 1166.1525,
|
| 239 |
+
"eval_samples_per_second": 14.781,
|
| 240 |
+
"eval_steps_per_second": 0.924,
|
| 241 |
+
"step": 16000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.91,
|
| 245 |
+
"learning_rate": 4.97795418450677e-05,
|
| 246 |
+
"loss": 1.1675,
|
| 247 |
+
"step": 16500
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.97,
|
| 251 |
+
"learning_rate": 4.976410006387394e-05,
|
| 252 |
+
"loss": 1.1585,
|
| 253 |
+
"step": 17000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.03,
|
| 257 |
+
"learning_rate": 4.974813818936012e-05,
|
| 258 |
+
"loss": 1.0833,
|
| 259 |
+
"step": 17500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.09,
|
| 263 |
+
"learning_rate": 4.973165655675605e-05,
|
| 264 |
+
"loss": 1.0134,
|
| 265 |
+
"step": 18000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.15,
|
| 269 |
+
"learning_rate": 4.9714655512207414e-05,
|
| 270 |
+
"loss": 1.0257,
|
| 271 |
+
"step": 18500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.2,
|
| 275 |
+
"learning_rate": 4.969713541276859e-05,
|
| 276 |
+
"loss": 1.0206,
|
| 277 |
+
"step": 19000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.26,
|
| 281 |
+
"learning_rate": 4.967909662639506e-05,
|
| 282 |
+
"loss": 1.0503,
|
| 283 |
+
"step": 19500
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.32,
|
| 287 |
+
"learning_rate": 4.9660539531935746e-05,
|
| 288 |
+
"loss": 1.0205,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.32,
|
| 293 |
+
"eval_bleu": 27.8937,
|
| 294 |
+
"eval_gen_len": 18.6638,
|
| 295 |
+
"eval_loss": 1.144053339958191,
|
| 296 |
+
"eval_runtime": 1147.812,
|
| 297 |
+
"eval_samples_per_second": 15.017,
|
| 298 |
+
"eval_steps_per_second": 0.939,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.38,
|
| 303 |
+
"learning_rate": 4.9641464519125e-05,
|
| 304 |
+
"loss": 1.032,
|
| 305 |
+
"step": 20500
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.44,
|
| 309 |
+
"learning_rate": 4.962187198857447e-05,
|
| 310 |
+
"loss": 1.0121,
|
| 311 |
+
"step": 21000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.49,
|
| 315 |
+
"learning_rate": 4.960176235176465e-05,
|
| 316 |
+
"loss": 1.013,
|
| 317 |
+
"step": 21500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.55,
|
| 321 |
+
"learning_rate": 4.958113603103627e-05,
|
| 322 |
+
"loss": 1.0169,
|
| 323 |
+
"step": 22000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.61,
|
| 327 |
+
"learning_rate": 4.9559993459581375e-05,
|
| 328 |
+
"loss": 1.0315,
|
| 329 |
+
"step": 22500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.67,
|
| 333 |
+
"learning_rate": 4.953833508143429e-05,
|
| 334 |
+
"loss": 1.0431,
|
| 335 |
+
"step": 23000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.73,
|
| 339 |
+
"learning_rate": 4.951616135146226e-05,
|
| 340 |
+
"loss": 1.0398,
|
| 341 |
+
"step": 23500
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 2.78,
|
| 345 |
+
"learning_rate": 4.949347273535588e-05,
|
| 346 |
+
"loss": 1.0217,
|
| 347 |
+
"step": 24000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.78,
|
| 351 |
+
"eval_bleu": 28.4149,
|
| 352 |
+
"eval_gen_len": 18.6666,
|
| 353 |
+
"eval_loss": 1.1301122903823853,
|
| 354 |
+
"eval_runtime": 1149.3346,
|
| 355 |
+
"eval_samples_per_second": 14.997,
|
| 356 |
+
"eval_steps_per_second": 0.938,
|
| 357 |
+
"step": 24000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.84,
|
| 361 |
+
"learning_rate": 4.9470269709619356e-05,
|
| 362 |
+
"loss": 1.0313,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.9,
|
| 367 |
+
"learning_rate": 4.944655276156047e-05,
|
| 368 |
+
"loss": 1.0127,
|
| 369 |
+
"step": 25000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.96,
|
| 373 |
+
"learning_rate": 4.9422322389280354e-05,
|
| 374 |
+
"loss": 1.0183,
|
| 375 |
+
"step": 25500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.02,
|
| 379 |
+
"learning_rate": 4.939757910166303e-05,
|
| 380 |
+
"loss": 0.9723,
|
| 381 |
+
"step": 26000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.07,
|
| 385 |
+
"learning_rate": 4.937232341836471e-05,
|
| 386 |
+
"loss": 0.8652,
|
| 387 |
+
"step": 26500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.13,
|
| 391 |
+
"learning_rate": 4.9346555869802904e-05,
|
| 392 |
+
"loss": 0.8811,
|
| 393 |
+
"step": 27000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.19,
|
| 397 |
+
"learning_rate": 4.9320276997145255e-05,
|
| 398 |
+
"loss": 0.8824,
|
| 399 |
+
"step": 27500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 3.25,
|
| 403 |
+
"learning_rate": 4.929348735229821e-05,
|
| 404 |
+
"loss": 0.8876,
|
| 405 |
+
"step": 28000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 3.25,
|
| 409 |
+
"eval_bleu": 28.5803,
|
| 410 |
+
"eval_gen_len": 18.6229,
|
| 411 |
+
"eval_loss": 1.1270363330841064,
|
| 412 |
+
"eval_runtime": 1153.2981,
|
| 413 |
+
"eval_samples_per_second": 14.946,
|
| 414 |
+
"eval_steps_per_second": 0.935,
|
| 415 |
+
"step": 28000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.31,
|
| 419 |
+
"learning_rate": 4.926618749789539e-05,
|
| 420 |
+
"loss": 0.8734,
|
| 421 |
+
"step": 28500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.36,
|
| 425 |
+
"learning_rate": 4.923837800728578e-05,
|
| 426 |
+
"loss": 0.8772,
|
| 427 |
+
"step": 29000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.42,
|
| 431 |
+
"learning_rate": 4.921005946452171e-05,
|
| 432 |
+
"loss": 0.8821,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 3.48,
|
| 437 |
+
"learning_rate": 4.918123246434658e-05,
|
| 438 |
+
"loss": 0.8774,
|
| 439 |
+
"step": 30000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 3.54,
|
| 443 |
+
"learning_rate": 4.915189761218237e-05,
|
| 444 |
+
"loss": 0.8932,
|
| 445 |
+
"step": 30500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 3.6,
|
| 449 |
+
"learning_rate": 4.912205552411688e-05,
|
| 450 |
+
"loss": 0.8745,
|
| 451 |
+
"step": 31000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.65,
|
| 455 |
+
"learning_rate": 4.9091706826890896e-05,
|
| 456 |
+
"loss": 0.8965,
|
| 457 |
+
"step": 31500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.71,
|
| 461 |
+
"learning_rate": 4.9060852157884893e-05,
|
| 462 |
+
"loss": 0.9024,
|
| 463 |
+
"step": 32000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.71,
|
| 467 |
+
"eval_bleu": 28.852,
|
| 468 |
+
"eval_gen_len": 18.7813,
|
| 469 |
+
"eval_loss": 1.118120551109314,
|
| 470 |
+
"eval_runtime": 1155.1982,
|
| 471 |
+
"eval_samples_per_second": 14.921,
|
| 472 |
+
"eval_steps_per_second": 0.933,
|
| 473 |
+
"step": 32000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.77,
|
| 477 |
+
"learning_rate": 4.9029492165105776e-05,
|
| 478 |
+
"loss": 0.8909,
|
| 479 |
+
"step": 32500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.83,
|
| 483 |
+
"learning_rate": 4.899762750717318e-05,
|
| 484 |
+
"loss": 0.8989,
|
| 485 |
+
"step": 33000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.89,
|
| 489 |
+
"learning_rate": 4.8965258853305676e-05,
|
| 490 |
+
"loss": 0.9054,
|
| 491 |
+
"step": 33500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.94,
|
| 495 |
+
"learning_rate": 4.893238688330673e-05,
|
| 496 |
+
"loss": 0.8852,
|
| 497 |
+
"step": 34000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.0,
|
| 501 |
+
"learning_rate": 4.889901228755036e-05,
|
| 502 |
+
"loss": 0.8879,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.06,
|
| 507 |
+
"learning_rate": 4.886513576696674e-05,
|
| 508 |
+
"loss": 0.7955,
|
| 509 |
+
"step": 35000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.12,
|
| 513 |
+
"learning_rate": 4.8830758033027385e-05,
|
| 514 |
+
"loss": 0.795,
|
| 515 |
+
"step": 35500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.18,
|
| 519 |
+
"learning_rate": 4.879587980773026e-05,
|
| 520 |
+
"loss": 0.7927,
|
| 521 |
+
"step": 36000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.18,
|
| 525 |
+
"eval_bleu": 28.3975,
|
| 526 |
+
"eval_gen_len": 18.4863,
|
| 527 |
+
"eval_loss": 1.1393358707427979,
|
| 528 |
+
"eval_runtime": 1137.7417,
|
| 529 |
+
"eval_samples_per_second": 15.15,
|
| 530 |
+
"eval_steps_per_second": 0.947,
|
| 531 |
+
"step": 36000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.23,
|
| 535 |
+
"learning_rate": 4.8760501823584607e-05,
|
| 536 |
+
"loss": 0.8031,
|
| 537 |
+
"step": 36500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 4.29,
|
| 541 |
+
"learning_rate": 4.8724624823595535e-05,
|
| 542 |
+
"loss": 0.7891,
|
| 543 |
+
"step": 37000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 4.35,
|
| 547 |
+
"learning_rate": 4.8688249561248456e-05,
|
| 548 |
+
"loss": 0.8028,
|
| 549 |
+
"step": 37500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 4.41,
|
| 553 |
+
"learning_rate": 4.8651376800493245e-05,
|
| 554 |
+
"loss": 0.8059,
|
| 555 |
+
"step": 38000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 4.47,
|
| 559 |
+
"learning_rate": 4.861400731572818e-05,
|
| 560 |
+
"loss": 0.8093,
|
| 561 |
+
"step": 38500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 4.52,
|
| 565 |
+
"learning_rate": 4.857614189178369e-05,
|
| 566 |
+
"loss": 0.8044,
|
| 567 |
+
"step": 39000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 4.58,
|
| 571 |
+
"learning_rate": 4.8537781323905896e-05,
|
| 572 |
+
"loss": 0.8098,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 4.64,
|
| 577 |
+
"learning_rate": 4.849892641773984e-05,
|
| 578 |
+
"loss": 0.8174,
|
| 579 |
+
"step": 40000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 4.64,
|
| 583 |
+
"eval_bleu": 28.6313,
|
| 584 |
+
"eval_gen_len": 18.3916,
|
| 585 |
+
"eval_loss": 1.124934434890747,
|
| 586 |
+
"eval_runtime": 1132.4078,
|
| 587 |
+
"eval_samples_per_second": 15.222,
|
| 588 |
+
"eval_steps_per_second": 0.952,
|
| 589 |
+
"step": 40000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 4.7,
|
| 593 |
+
"learning_rate": 4.845957798931265e-05,
|
| 594 |
+
"loss": 0.821,
|
| 595 |
+
"step": 40500
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.76,
|
| 599 |
+
"learning_rate": 4.841973686501636e-05,
|
| 600 |
+
"loss": 0.8104,
|
| 601 |
+
"step": 41000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 4.81,
|
| 605 |
+
"learning_rate": 4.8379403881590544e-05,
|
| 606 |
+
"loss": 0.8155,
|
| 607 |
+
"step": 41500
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 4.87,
|
| 611 |
+
"learning_rate": 4.833857988610474e-05,
|
| 612 |
+
"loss": 0.8213,
|
| 613 |
+
"step": 42000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 4.93,
|
| 617 |
+
"learning_rate": 4.8297265735940714e-05,
|
| 618 |
+
"loss": 0.819,
|
| 619 |
+
"step": 42500
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 4.99,
|
| 623 |
+
"learning_rate": 4.825546229877439e-05,
|
| 624 |
+
"loss": 0.8255,
|
| 625 |
+
"step": 43000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 5.05,
|
| 629 |
+
"learning_rate": 4.8213170452557655e-05,
|
| 630 |
+
"loss": 0.7598,
|
| 631 |
+
"step": 43500
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 5.11,
|
| 635 |
+
"learning_rate": 4.817039108549991e-05,
|
| 636 |
+
"loss": 0.7434,
|
| 637 |
+
"step": 44000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.11,
|
| 641 |
+
"eval_bleu": 28.2898,
|
| 642 |
+
"eval_gen_len": 18.7739,
|
| 643 |
+
"eval_loss": 1.1695784330368042,
|
| 644 |
+
"eval_runtime": 1160.4522,
|
| 645 |
+
"eval_samples_per_second": 14.854,
|
| 646 |
+
"eval_steps_per_second": 0.929,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 5.16,
|
| 651 |
+
"learning_rate": 4.812712509604945e-05,
|
| 652 |
+
"loss": 0.7564,
|
| 653 |
+
"step": 44500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 5.22,
|
| 657 |
+
"learning_rate": 4.808337339287453e-05,
|
| 658 |
+
"loss": 0.7496,
|
| 659 |
+
"step": 45000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 5.28,
|
| 663 |
+
"learning_rate": 4.803913689484436e-05,
|
| 664 |
+
"loss": 0.7381,
|
| 665 |
+
"step": 45500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.34,
|
| 669 |
+
"learning_rate": 4.7994416531009745e-05,
|
| 670 |
+
"loss": 0.7394,
|
| 671 |
+
"step": 46000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 5.4,
|
| 675 |
+
"learning_rate": 4.794921324058358e-05,
|
| 676 |
+
"loss": 0.7468,
|
| 677 |
+
"step": 46500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.45,
|
| 681 |
+
"learning_rate": 4.790352797292116e-05,
|
| 682 |
+
"loss": 0.7448,
|
| 683 |
+
"step": 47000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.51,
|
| 687 |
+
"learning_rate": 4.785736168750022e-05,
|
| 688 |
+
"loss": 0.7339,
|
| 689 |
+
"step": 47500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.57,
|
| 693 |
+
"learning_rate": 4.781071535390078e-05,
|
| 694 |
+
"loss": 0.7416,
|
| 695 |
+
"step": 48000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.57,
|
| 699 |
+
"eval_bleu": 28.9507,
|
| 700 |
+
"eval_gen_len": 18.6744,
|
| 701 |
+
"eval_loss": 1.1450848579406738,
|
| 702 |
+
"eval_runtime": 1148.6011,
|
| 703 |
+
"eval_samples_per_second": 15.007,
|
| 704 |
+
"eval_steps_per_second": 0.939,
|
| 705 |
+
"step": 48000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 5.63,
|
| 709 |
+
"learning_rate": 4.7763589951784795e-05,
|
| 710 |
+
"loss": 0.7509,
|
| 711 |
+
"step": 48500
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 5.69,
|
| 715 |
+
"learning_rate": 4.771598647087556e-05,
|
| 716 |
+
"loss": 0.7383,
|
| 717 |
+
"step": 49000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 5.74,
|
| 721 |
+
"learning_rate": 4.7667905910936926e-05,
|
| 722 |
+
"loss": 0.735,
|
| 723 |
+
"step": 49500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 5.8,
|
| 727 |
+
"learning_rate": 4.761934928175236e-05,
|
| 728 |
+
"loss": 0.7541,
|
| 729 |
+
"step": 50000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 5.86,
|
| 733 |
+
"learning_rate": 4.7570317603103634e-05,
|
| 734 |
+
"loss": 0.7554,
|
| 735 |
+
"step": 50500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 5.92,
|
| 739 |
+
"learning_rate": 4.75208119047495e-05,
|
| 740 |
+
"loss": 0.7511,
|
| 741 |
+
"step": 51000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 5.98,
|
| 745 |
+
"learning_rate": 4.747083322640401e-05,
|
| 746 |
+
"loss": 0.7599,
|
| 747 |
+
"step": 51500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 6.03,
|
| 751 |
+
"learning_rate": 4.742038261771472e-05,
|
| 752 |
+
"loss": 0.689,
|
| 753 |
+
"step": 52000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 6.03,
|
| 757 |
+
"eval_bleu": 28.3532,
|
| 758 |
+
"eval_gen_len": 18.4481,
|
| 759 |
+
"eval_loss": 1.1758701801300049,
|
| 760 |
+
"eval_runtime": 1136.8302,
|
| 761 |
+
"eval_samples_per_second": 15.162,
|
| 762 |
+
"eval_steps_per_second": 0.948,
|
| 763 |
+
"step": 52000
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 6.09,
|
| 767 |
+
"learning_rate": 4.73694611382406e-05,
|
| 768 |
+
"loss": 0.6386,
|
| 769 |
+
"step": 52500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 6.15,
|
| 773 |
+
"learning_rate": 4.7318069857429815e-05,
|
| 774 |
+
"loss": 0.6589,
|
| 775 |
+
"step": 53000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 6.21,
|
| 779 |
+
"learning_rate": 4.7266209854597246e-05,
|
| 780 |
+
"loss": 0.7347,
|
| 781 |
+
"step": 53500
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 6.27,
|
| 785 |
+
"learning_rate": 4.721388221890185e-05,
|
| 786 |
+
"loss": 0.7288,
|
| 787 |
+
"step": 54000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 6.32,
|
| 791 |
+
"learning_rate": 4.716108804932374e-05,
|
| 792 |
+
"loss": 0.7354,
|
| 793 |
+
"step": 54500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 6.38,
|
| 797 |
+
"learning_rate": 4.7107828454641146e-05,
|
| 798 |
+
"loss": 0.7334,
|
| 799 |
+
"step": 55000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 6.44,
|
| 803 |
+
"learning_rate": 4.7054104553407104e-05,
|
| 804 |
+
"loss": 0.7244,
|
| 805 |
+
"step": 55500
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 6.5,
|
| 809 |
+
"learning_rate": 4.699991747392598e-05,
|
| 810 |
+
"loss": 0.7238,
|
| 811 |
+
"step": 56000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 6.5,
|
| 815 |
+
"eval_bleu": 28.3827,
|
| 816 |
+
"eval_gen_len": 18.7038,
|
| 817 |
+
"eval_loss": 1.1825261116027832,
|
| 818 |
+
"eval_runtime": 1155.4906,
|
| 819 |
+
"eval_samples_per_second": 14.917,
|
| 820 |
+
"eval_steps_per_second": 0.933,
|
| 821 |
+
"step": 56000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 6.56,
|
| 825 |
+
"learning_rate": 4.6945268354229766e-05,
|
| 826 |
+
"loss": 0.7205,
|
| 827 |
+
"step": 56500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 6.61,
|
| 831 |
+
"learning_rate": 4.689015834205418e-05,
|
| 832 |
+
"loss": 0.7157,
|
| 833 |
+
"step": 57000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 6.67,
|
| 837 |
+
"learning_rate": 4.6834588594814556e-05,
|
| 838 |
+
"loss": 0.7078,
|
| 839 |
+
"step": 57500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.73,
|
| 843 |
+
"learning_rate": 4.677856027958154e-05,
|
| 844 |
+
"loss": 0.71,
|
| 845 |
+
"step": 58000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 6.79,
|
| 849 |
+
"learning_rate": 4.6722074573056576e-05,
|
| 850 |
+
"loss": 0.7082,
|
| 851 |
+
"step": 58500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 6.85,
|
| 855 |
+
"learning_rate": 4.6665132661547215e-05,
|
| 856 |
+
"loss": 0.7258,
|
| 857 |
+
"step": 59000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 6.9,
|
| 861 |
+
"learning_rate": 4.660773574094218e-05,
|
| 862 |
+
"loss": 0.7306,
|
| 863 |
+
"step": 59500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 6.96,
|
| 867 |
+
"learning_rate": 4.6549885016686236e-05,
|
| 868 |
+
"loss": 0.7238,
|
| 869 |
+
"step": 60000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.96,
|
| 873 |
+
"eval_bleu": 28.8248,
|
| 874 |
+
"eval_gen_len": 18.5073,
|
| 875 |
+
"eval_loss": 1.1675989627838135,
|
| 876 |
+
"eval_runtime": 1137.8605,
|
| 877 |
+
"eval_samples_per_second": 15.149,
|
| 878 |
+
"eval_steps_per_second": 0.947,
|
| 879 |
+
"step": 60000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 7.02,
|
| 883 |
+
"learning_rate": 4.649158170375489e-05,
|
| 884 |
+
"loss": 0.681,
|
| 885 |
+
"step": 60500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 7.08,
|
| 889 |
+
"learning_rate": 4.6432827026628894e-05,
|
| 890 |
+
"loss": 0.571,
|
| 891 |
+
"step": 61000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 7.14,
|
| 895 |
+
"learning_rate": 4.63736222192685e-05,
|
| 896 |
+
"loss": 0.5642,
|
| 897 |
+
"step": 61500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 7.19,
|
| 901 |
+
"learning_rate": 4.631396852508754e-05,
|
| 902 |
+
"loss": 0.5604,
|
| 903 |
+
"step": 62000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 7.25,
|
| 907 |
+
"learning_rate": 4.625386719692734e-05,
|
| 908 |
+
"loss": 0.5799,
|
| 909 |
+
"step": 62500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 7.31,
|
| 913 |
+
"learning_rate": 4.61933194970304e-05,
|
| 914 |
+
"loss": 0.5989,
|
| 915 |
+
"step": 63000
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 7.37,
|
| 919 |
+
"learning_rate": 4.613232669701384e-05,
|
| 920 |
+
"loss": 0.6339,
|
| 921 |
+
"step": 63500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 7.43,
|
| 925 |
+
"learning_rate": 4.607089007784279e-05,
|
| 926 |
+
"loss": 0.657,
|
| 927 |
+
"step": 64000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 7.43,
|
| 931 |
+
"eval_bleu": 27.4378,
|
| 932 |
+
"eval_gen_len": 18.4196,
|
| 933 |
+
"eval_loss": 1.2514448165893555,
|
| 934 |
+
"eval_runtime": 1132.4116,
|
| 935 |
+
"eval_samples_per_second": 15.221,
|
| 936 |
+
"eval_steps_per_second": 0.952,
|
| 937 |
+
"step": 64000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 7.43,
|
| 941 |
+
"step": 64000,
|
| 942 |
+
"total_flos": 2.2192449043268895e+18,
|
| 943 |
+
"train_loss": 0.9512434206008911,
|
| 944 |
+
"train_runtime": 65044.6534,
|
| 945 |
+
"train_samples_per_second": 84.803,
|
| 946 |
+
"train_steps_per_second": 5.3
|
| 947 |
+
}
|
| 948 |
+
],
|
| 949 |
+
"logging_steps": 500,
|
| 950 |
+
"max_steps": 344760,
|
| 951 |
+
"num_train_epochs": 40,
|
| 952 |
+
"save_steps": 4000,
|
| 953 |
+
"total_flos": 2.2192449043268895e+18,
|
| 954 |
+
"trial_name": null,
|
| 955 |
+
"trial_params": null
|
| 956 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:812228f1e0a1b305a43ac2ec4cf6a158a90cadcdd15b8aec2eea894cd17b40de
|
| 3 |
+
size 4728
|